現行の自発性早産予測バイオマーカーの性能特性

Preterm birth (PTB), or birth occurring before 37 weeks' gestation, remains a significant public health burden, accounting for 10% of live births annually in the United States and incurring substantial healthcare expenditures. Our understanding of the molecular mechanisms underlying spontaneous preterm birth (sPTB) has advanced across the previous 4 decades, yet precise prediction tools and prevention strategies are lacking.

Numerous studies have identified potential anatomical and molecular risk factors for sPTB, including sonographic characteristics of the cervix; maternal serum circulating RNA and proteins; maternal urine metabolic byproducts; cervicovaginal cytokine, microbiome, and metabolome composition; amniotic fluid cytokines; umbilical cord blood leukocyte DNA methylation status; and placental transcriptome profiles. This review focuses on recent developments in sPTB biomarker determination among singleton gestations.

Herein, we synthesize and evaluate the test characteristics of candidate biomarkers of sPTB, concluding that no single biomarker can accurately predict sPTB. However, several individual or combined panels of biomolecules, including some commercially available, carry clinically significant predictive information. These biomarkers include cervical ultrasonography, the ratio of insulin-like growth factor-binding protein 4 to sex-hormone binding globulin, panels of urinary metabolites and amniotic fluid proteins, and maternal circulating cell-free RNA. Future integration of select biomarkers drawn from prospective validation cohorts into existing risk stratification strategies may enhance sPTB prediction, thereby identifying patients at greatest risk.